1. Field of the Invention
The present invention relates to an image processing method for discriminating a scene of an input image and performing image processing according to the scene thus discriminated, an image processing apparatus thereof, and an imaging apparatus thereof.
2. Description of the Related Art
Conventionally, there is an image processing apparatus that discriminates a scene of an image or a type of an object according to the scene (i.e., the type of the object) thus discriminated.
For example, Japanese Patent Laid-open Publication No. 2000-259372 discloses an image output apparatus that discriminates whether the input image data is a vivid color scene (i.e., a vivid object) and outputs the image data at high resolution in a case where the image data is the vivid color scene.
In the conventional technique disclosed in the above described Japanese Patent Laid-open Publication No. 2000-259372, whether the image data is the vivid color scene is determined according to whether the number of pixels having a high chroma is more than a predetermined threshold or less than the predetermined threshold.
However, in the conventional method that discriminates the vivid color scene based only on chroma information of the image, a discrimination result of the scene sometimes does not match a feeling of a user, i.e., a feeling of a human.
For example, in a case where an image includes the many number of colors, the user tends to view the image as the vivid color scene even if the chroma of each of the colors is low, in comparison with a case that the image includes less colors.
There is a case where the user does not intend to capture a vivid color scene even if the image includes a lot of high chroma pixels. An example of such case is a case that an object is captured with a background, e.g., with a wall behind an object, that is uniformly painted with a high chroma color. As described above, even with the same chroma, there is a case where the user tends to view the image as the vivid color scene whereas there is a case where the user does not view the image as the vivid color scene. In a case as described above, it is hard to discriminate the scene according to the user's views or the user's intention according to the conventional discrimination method.